OpenAI's Codex now encrypts instructions between AI agents, leaving developers blind to internal delegation

OpenAI’s Codex now encrypts instructions between its AI agents, replacing readable task descriptions with unreadable strings and preventing developers from tracking internal delegation. The encryption is mandatory for larger GPT-5.6 models Sol and Terra, though users report occasional failures due to decryption issues. OpenAI has not explained the change, which may aim to prevent rivals from training on internal communications or simply align with existing API encryption practices.